Networks that Learn: Towards Autonomous Network Systems

Dear Colleagues,

Autonomous systems are becoming wide spread: autonomous cars will soon be a reality, smartphones are already able to recognize our voice and our face and many health applications are leveraging machine learning tools, e.g., to automate the diagnosis of diseases. It is undeniable that the advent of deep learning, supported by an unprecedented amount of computational power, has paved the way for extremely complex tasks to be delegated to machines. These machines or artificial intelligences use any sort of input to obtain the desired output autonomously, building decision making solutions that often defy the understanding of the researchers who programmed them.

Within the telecommunications field, machine learning techniques are attracting an increasing interest too, especially to provide networks with the ability to learn and adapt their parameters, or the network protocols in use, taking traffic patterns, application and users' needs into account. Despite this interest, the application of learning techniques to the telecommunications domain is still in its infancy and no comprehensive solution has been proposed so far. In contrast to image recognition, which is a rather mature field, identifying modulation, coding schemes, protocols, traffic load patterns and other communication features has been elusive so far. Nonetheless, we believe that communications networks can and should also be managed as any other sensor system, i.e., sensing the environment, making decisions, and acting upon.

With this special issue, our interest is centered around "networks that learn", where artificial intelligence tools are applied to the optimization of communication networks. We believe that learning and adaptation will be key to achieve the maximum level of spectrum efficiency, by letting competing systems to autonomously cooperate with one another, simultaneously understanding the surrounding environment and adapting to it. Along these lines, a relevant initiative is DARPA's Spectrum Collaboration Challenge, which aims at exploiting machine learning to overcome the scarcity in the radio frequency spectrum and is pushing towards a so called network autonomy.

In conclusion, we encourage the submission of papers on the latest advances on machine learning applied to communication systems. In particular, we warmly welcome submissions addressing, but not limited to, the following topics:

• Autonomous network optimization through machine learning
• Radio frequency characteristics identification and fingerprinting
• Traffic pattern analysis
• Blind spectrum cooperation strategies (i.e., no primary network is known in advance)

Prof. Michele Rossi
Dr. Nicola Bui
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Journal of Sensor and Actuator Networks is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2000 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• network autonomy
• machine learning
• cognitive networks
• cooperative spectrum